[Talk-scotland] Mapping of individual trees versus wooded areas

Eric Grosso eric.grosso.os at gmail.com
Tue Jan 17 15:57:05 UTC 2017


Hi Donald,

I'm reading your message just now so sorry to have not answered before. I
read also the discussion in the note you talk about.

Thanks a lot for your message and to raise this problem.

I totally agree with what you wrote. We (MESH) tried to do our best to keep
most of the woodlands when we had the feeling, obviously not only from the
satellite imagery but from our multiple surveys, that there were enough
trees on the ground to constitute a woodland. We are very happy to discuss
any of our changes and if some of them are problematic, we'll change them
back as they were before -- using the "revert changeset" tool partially to
bring back the former woodlands. Sorry about that as we should have engaged
more with you guys in Edinburgh, via OSM notes for example.

The point that you make here, Donald, about the rendering of the woodland
is a very good one. A good compromise could be to draw both of them
(woodland and individual trees) when necessary.

It's always difficult to judge if a bunch of trees can be considered as a
woodland or not, as everyone has a different feeling about what a woodland
is. For example, can the trees in the Meadows could be considered as
woodland? Personally, I don't think so as I considered trees being part of
tree rows, except the two bits in the West and at the bottom of Middle
Meadow Walk, but even these two can be considered as open spaces. But I'm
pretty sure that someone else has a different opinion on this.

Finally, the main problem mapping individual trees and woodlands, it's how
big a single tree can be seen from the Bing satellite imagery ("canopy"
point of view) vs from the ground ("trunk" point of view) -- the Holyrood
Palace gardens are a quite good example of that. Sometimes, adding
individual trees can quickly become an addiction and that's probably why we
over-mapped them in some places.

I'll be tonight at the pub meeting and I'll be very happy to discuss it. If
needed or asked, we'll make the required changes this week. If you can't
attend tonight, just let us know via a note or via this discussion. Thanks
in advance.

Cheers,
Eric


On 18 December 2016 at 11:51, Donald Noble <drnoble at gmail.com> wrote:

> I noticed on the map recently that there are further areas of woodland
> that have been deleted and replaced with a lot of individual trees. [1]
> There was a previous discussion regarding this subject at
> http://www.openstreetmap.org/note/733730
>
> I still hold by my view that mapping individual trees/groups of trees in
> gardens looks great (and thanks to the MESH team for their hard work on
> this) but larger areas of mixed woodland shouldn't be represented in this
> way. Many of these areas will have a complex understory of shrubs and new
> growth. It can also be difficult to pick out/separate individual tree
> crowns from aerial imagery which might be several years out of date.
>
> There is no way to aggregate areas of individually mapped trees and show
> them over a bigger area. Woodland is mapped at zoom level 8, whereas trees
> don't appear on z15, only at z16 and above.
>
> I don't want to get into an edit war over this, but I thought some sort of
> consensus had been reached after the last discussion. My view is only that,
> and if other people feel the opposite please say.
>
> regards, Donald
>
> [1] The areas I noticed were around Corsorphine, along the railway path,
> but I can't find a changeset where the woodland was deleted
>
> _______________________________________________
> Talk-scotland mailing list
> Talk-scotland at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-scotland
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-scotland/attachments/20170117/a7213343/attachment.html>


More information about the Talk-scotland mailing list